Phoenix 2007 (new thread)

Quoting Smileman

Disliked

Hi Hendrick
I know your are very busy in these days and I don't want to bother you to much, but I'm very eager waiting for explanation you will give us about how can anyone find the Phoenix settings for the next coming months?

Thanks you very much for your help

Ignored

Quoting Capt_Oliv

Disliked

Hi Hendrick,

Many thanks for sharing with us your Phoenix EA. It is great and I wish you all the best for the Automated Trading Championship 2006.

As I looked into the code to try to find out it's "magic formla", I was not sure if the divergence function was coded as originally intended : I refering to the following part

" dv2 = ((maF1 - maS1) - (maF2 - maS2));
//----
return(dv1 - dv2);"

I would guess it was intended as :

" dv2 = maF2 - maS2;
//----
return(dv1 - dv2);"

Sorry in advance, if this topic as already been addressed in this thread, I just didn't managed to find the it.

Olivier

Ignored

Quoting athalon7

Disliked

Here is the profit factor/net profit totals for 5 popular currencies. Hendrick says we should look every two weeks to see if tweeking is necessary, so I based the numbers on two week differentials. I could not Upload the spreadsheet as .xls files are not permitted.

Now the questions: how much does it have to drop before a tweak is necessary and where do we start with the tweaking?

I noticed that the the GBP/USD is the only currency that has the OSMA Fast number higher than the OSMA Slow for what it is worth.

Hang in there on the trades. This is a marathon, not a sprint. Easy to say, hard to do I know, but the backtesting proves that bad trades will be made. If this wasn't the case there wouldn't be a market.

Ignored

Quoting Palmer

Disliked

Harry:

OK, I was thinking about the logic wrong. But yes, it would have to be forward tested, not backtested.

Without logic there is no reason...

Ignored

Quoting Hendrick

Disliked

I would be very grateful if some of you can make the next summary:<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

<o></o>

Look at your Account History and look for every trade at its opening and closing time. Is there some kind of correlation between the total time a trade is open and if it's profitable or not? I've not analysed my own Account History but it seems to me that the longer a trade is open the less chance it will be profitable.<o></o>

<o></o>

So, for those who have an Account History of 2 weeks or more and if you have the time can you please make such a summary. It would really help me!<o></o>

<o></o>

Ignored

Quoting 1trader2

Disliked

Hendrick,

I have been trading Phoenix since 9/4/06, updating versions as you post them. Have had 63 total trades, 51 profitable, 12 losses. Days to loss as follows:

Open Close Days Pair
9/15 9/15 <1 usdjpy
9/25 9/26 1 eurjpy
9/26 9/27 1 usdjpy
9/27 9/28 1 eurjpy
9/27 9/29 2 usd/jpy
10/6 10/6 <7 hrs usd/jpy
10/6 10/9 3 gbp/jpy
10/9 10/10 1 usd/cad
10/9 10/11 2 eur/usd
10/6 10/11 5 usd/jpy
10/9 10/16 7 eur/jpy
10/17 10/17 <5 hrs gbp/jpy


My observation does not substantiate what you suspected. What I see is 80% of the losses are in the JPY pairs. 10 of the losing pairs were attributed to JPY. Total loss from these pairs was: $1730-776 pips. However 30 of the 51 profitable trades (59%) were in the JPY pairs. I use FXDD and trade the 8 pairs you recommended, half of which are the JPY pairs. Any recommendations?

1Trader2

Ignored

Quoting Hendrick

Disliked

Hi Olivier,

As far as I understood there are indeed two versions of Diverge. When I have the time I will test both versions on Phoenix. Although I don't think it will make a lot of difference. Thanks Olivier for this contribution.

Ignored

Quoting Palmer

Disliked

Is it just me or does the class see something here??? Buehler....Buehler...?

Ignored

Quoting athalon7

Disliked

Hendrick-

Have you persued a "dynamic stop"? See the enclosed Bitmap, it should be
self explanatory.

Ignored

Quoting Capt_Oliv

Disliked

Hi Hendrick,

Well no, what I was trying to gently say is : I think there is a bug in the divergence function of Phoenix_EA_v4_2_03.mq4 (which give great result BTW)

As it is today in P4 :

"double divergence(int F_Period, int S_Period, int F_Price, int S_Price, int mypos)
{
int i;
//----
double maF1, maF2, maS1, maS2;
double dv1, dv2;
//----
maF1 = iMA(Symbol(), 0, F_Period, 0, MODE_SMA, F_Price, mypos);
maS1 = iMA(Symbol(), 0, S_Period, 0, MODE_SMA, S_Price, mypos);
dv1 = (maF1 - maS1);

maF2 = iMA(Symbol(), 0, F_Period, 0, MODE_SMA, F_Price, mypos + 1);
maS2 = iMA(Symbol(), 0, S_Period, 0, MODE_SMA, S_Price, mypos + 1);
dv2 = ((maF1 - maS1) - (maF2 - maS2));
//----
return(dv1 - dv2);
}"

the divergence function is not computing a divergence, it returns :

dv1 - dv2 = (maF1 - maS1) - ((maF1 - maS1) - (maF2 - maS2)) = maF2 - maS2

so either,
it is useless to compute maF1 and maS1
or dv2 should be equal to (maF2 - maS2) to compute a divergence (which, I think was your original intention)

This "correction" may improve results, but most certainly will require new
prefered settings tweaking, which take some time...

Olivier

Ignored